Downlink Multi-user Hybrid Beamforming For Millimeter Wave Massive MIMO System

The fifth generation wireless communication systems can provide Gbps data transfer rates due to the large bandwidth of the millimeter-wave(mmWave)frequency and the high gain of Massive Multiple-Input Multiple-Output(MIMO)systems.The most significant features of mmWave are wide band and short wavelengths,effective use of mmWave bandwidth resources will greatly ease the problem of spectrum resource constraints.However,mmWave in the propagation process will have a serious path loss,while large-scale antennas high array gain just can make up for this.The main feature of the large-scale MIMO system is the configuration of a large-scale antennas,the number of antennas configured on the base station(BS)has reached several hundred or even thousands,these antennas are placed together to form a linear array or a planar array,the array gain of these antennas is concentrated within a very small range through beamforming technology,to achieve accurate space division multiplexing,and space degrees of freedom are greatly increased,thus the system capacity and spectral efficiency will be greatly improved.Although the array gain of large-scale antennas can effectively compensate for the loss of performance due to large path losses in mmWave communications.A large number of antenna arrays will result in high cost,high computational complexity,and high energy consumption.Hybrid analog/digital beamforming can effectively reduce the number of Radio Frequency(RF)links and reduce computational complexity.And multi-user Space Division Multiplexing(SDM)in millimeter-wave systems can achieve higher rates,and further improve the system capacity.This paper focuses on the precoding algorithm of multi-user large-scale MIMO system in millimeter-wave communication.The principles of Massive MIMO system and mmWave Massive MIMO system are analyzed in detail.In addition,several traditional algorithms are also studied,and the shortcomings in the existing algorithms are pointed out,then an effective analog/digital hybrid beamforming algorithm for multi-user mmWave Massive MIMO system is proposed.In order to make better use of directional gain of analog beamforming and flexible baseband processing of digital beamforming,this paper puts forward to an approach of using the analog beamforming to select best analog beams to maximize the power of the target user and minimize the interference to other users.In addition,the digital beamforming is designed to maximize the energy efficiency of the target user,and the optimization problem is solved by zero-gradient-based method.Simulation results show that the algorithm has better bit error rate(BER)performance compared with the traditional hybrid beamforming and with the number of users increases the sum rate of system significantly improved.The study of this paper proves that multiuser MIMO(MU-MIMO)can be used as a perfect candidate for mmWave Massive MIMO communication systems.In addition,the optimal beam selection of analog beamforming can more effectively suppress inter-user interference,while the digital beamforming can greatly improve system performance through flexible baseband processing.